Introduction to visibility tests

Before we do anything else, we 'd all like to thank you for taking time to help us by participating. To get you started, you will either have one of us with you, or immediately available by telephone , depending on your past experience with the X-window environment and the images involved. But before that happens, we'd like to give everyone some common background.

Our goal in this project is to measure the visible effects of a recently-implemented image processing technique when applied to a variety of medical images, as a function of the degree & kind of processing. We thus want to find the range of processing resulting in images which are (for all practical purposes) either visibly indistinguishable or superior to unprocessed images, or no worse than unprocessed images. (You might say we are looking for the cosmetic and greatest-tolerable-degradation levels of processing.) Initially, we will concentrate on computed-tomography (CT) scans of the liver & lung, and plan to examine computed radiography images as they become available. You will be asked to perform the following tasks:

  1. point out the positions or absence of abnormalities in an image or a set of images.
  2. classify the relationship of two images.

We welcome your comments regarding the testing procedures, images, and so forth at any time; comments about a particular image or set of images should be made while viewing them (during a test segment) by clicking on the Comment button, so that we can coordinate the comment with the image. Though the test need not be completed within any particular time limit, we would prefer that you be free of distractions such as the telephone so that you can commit your full attention to evaluating the images. If you're unable to finish without interruption, you can finish at another time by clicking the Postpone button.

The tasks are detailed beginning with the next page....

Table of Contents


This document generated by SpHyDir, another fine product of PC Lube and Tune.